CN116933201A - Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile - Google Patents
Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile Download PDFInfo
- Publication number
- CN116933201A CN116933201A CN202310817174.6A CN202310817174A CN116933201A CN 116933201 A CN116933201 A CN 116933201A CN 202310817174 A CN202310817174 A CN 202310817174A CN 116933201 A CN116933201 A CN 116933201A
- Authority
- CN
- China
- Prior art keywords
- data
- charging pile
- risk
- low
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000004140 cleaning Methods 0.000 claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims description 35
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 11
- 238000012502 risk assessment Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000005429 filling process Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 34
- 238000012545 processing Methods 0.000 description 14
- 238000004458 analytical method Methods 0.000 description 11
- BUGBHKTXTAQXES-UHFFFAOYSA-N Selenium Chemical compound [Se] BUGBHKTXTAQXES-UHFFFAOYSA-N 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 229910052711 selenium Inorganic materials 0.000 description 5
- 239000011669 selenium Substances 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 238000009795 derivation Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000009960 carding Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
- G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for identifying the illegal electricity consumption behavior of a low-voltage charging pile, comprising the following steps: s1: collecting original data; s2: cleaning the original data to obtain cleaned data; s3: extracting characteristic parameters based on the cleaned data, and setting assessment factors based on the characteristic parameters; s4: obtaining a risk score of the assessment factor based on the characteristic parameters, and obtaining a comprehensive risk score based on the risk score; s5: determining offending user information based on the risk score and the comprehensive risk score, and outputting a power utilization curve graph based on the offending user information; s6: the default electrical behavior is identified based on the electrical usage graph. The beneficial effects of the invention are as follows: the recognition accuracy of the illegal electricity utilization behavior of the low-voltage charging pile can be improved.
Description
Technical Field
The invention relates to the technical field of electricity analysis of charging piles, in particular to a method and a system for identifying illegal electricity utilization behaviors of low-voltage charging piles.
Background
With the rapid growth of electric vehicles and charging piles, partial users bring pressure to the power grid during the excessive charging of late night peak, and the economic loss of the power grid is caused, so that the illegal electricity utilization behavior of the users needs to be identified.
In the prior art, the default electricity consumption is generally judged according to the charge amount of the charging pile to the electric automobile, the complex default electricity consumption behavior is difficult to accurately identify, and the problem that the identification accuracy of the default electricity consumption behavior of the low-voltage charging pile cannot be improved exists under the condition of false alarm or missing alarm.
For example, a "cell iris recognition electric vehicle charging pile" disclosed in chinese patent literature, its bulletin number: CN107424318A, filing date: in 2017, on the day of 07 and 31, when the electric vehicle is in a charging process, the electric quantity detection circuit monitors the electric quantity of the storage battery of the electric vehicle, and can timely close the current charging pile when the charging is completed, so that the storage battery is protected, the energy consumption is saved, but the problem that the recognition accuracy of the illegal electricity consumption of the low-voltage charging pile cannot be improved exists.
Disclosure of Invention
Aiming at the defect that the identification accuracy of the illegal electricity consumption behavior of the low-voltage charging pile cannot be improved in the prior art, the invention provides a method and a system for identifying the illegal electricity consumption behavior of the low-voltage charging pile, which can improve the identification accuracy of the illegal electricity consumption behavior of the low-voltage charging pile.
The technical scheme of the invention is that the method for identifying the illegal electricity consumption behavior of the low-voltage charging pile comprises the following steps:
s1: collecting original data;
s2: cleaning the original data to obtain cleaned data;
s3: extracting characteristic parameters based on the cleaned data, and setting assessment factors based on the characteristic parameters;
s4: obtaining a risk score of the assessment factor based on the characteristic parameters, and obtaining a comprehensive risk score based on the risk score;
s5: determining offending user information based on the risk score and the comprehensive risk score, and outputting a power utilization curve graph based on the offending user information; s6: the default electrical behavior is identified based on the electrical usage graph.
In the scheme, load data and user data are collected, the load data and the user data are spliced into original data, cleaning processes such as denoising and abnormal value deleting are carried out on the original data, cleaned data are output, a plurality of characteristic parameters are extracted based on the cleaned data, risk scores of all assessment factors are obtained based on the data of the plurality of characteristic parameters, comprehensive risk scores are obtained through the risk scores, the user numbers of the offending users are judged and output based on the comprehensive risk scores and the risk scores, an original load data curve of the offending users in one week is read and drawn based on the user numbers of the offending users, an electricity consumption curve is output, the offending electricity consumption behavior is judged based on the electricity consumption curve, and then relevant processing is carried out. The recognition accuracy of the illegal electricity utilization behavior of the low-voltage charging pile can be improved.
Preferably, in S1, the original data includes load data and user data.
In the scheme, the load data is used for carrying out electricity analysis, the user data is used for determining the user carrying out electricity analysis, and the electricity load of the appointed user can be determined, so that the target user is determined after the illegal electricity consumption behavior is identified, and then related processing is carried out.
Preferably, in S2, the cleaning process includes a denoising process, a null value filling process, and an outlier deletion process.
In the scheme, data denoising processing is carried out on users with load data, and the original data with the reject power smaller than the reject power threshold is rejected, wherein the reject power threshold is 20W. And filling the missing data in a short time with the missing value by using the latest non-zero sampling point data. The common standby power of the low-voltage charging pile is generally smaller than 0.01kw, the default power of the data preprocessing algorithm is set to be in a charging state when the default power is larger than a charging power threshold value, and the charging power threshold value is 0.02kw.
Preferably, in S3, the characteristic parameters include average power, variance, peak-to-valley ratio, integrated electric quantity, maximum load, minimum load, charging pile usage rate and number of load abrupt changes, and the evaluation factors include variance factor, duty factor, load segment number factor, average power factor at charging, charging pile usage rate factor, integrated back-calculation electric quantity factor and capacity usage rate factor.
In the scheme, the risk score of the assessment factor is calculated through the characteristic parameters of the electric load, and the variance factor reflects the overall fluctuation condition during charging, so that the larger the variance is, the larger the fluctuation is and the higher the risk is; the higher the duty cycle ratio, the higher the risk; the more the number of load segments is, the higher the risk is; average power factor at charge, generally 6.5kw-7.5kw risk is lower; the higher the utilization factor of the charging pile is, the higher the risk is; the electric quantity factor is calculated in an integral back-calculation mode, and the higher the electric quantity is, the higher the risk is.
Preferably, in S4, the assessment factors and the risk scores are in one-to-one correspondence, and the composite risk score is equal to the product of the risk scores of some or all of the assessment factors.
Preferably, S5 comprises the steps of:
s51: setting a risk assessment range;
s52: judging whether the risk assessment range corresponding to the comprehensive risk score is high risk, if so, performing step S53; if not, judging that the user is not default;
s53: judging whether the risk scores of a plurality of and more than one assessment factors are larger than or equal to a risk threshold value; if yes, judging as a default user, outputting a user number of the default user, reading and drawing a load data curve within the appointed time of the default user based on the user number, and outputting an electricity utilization curve; if not, judging that the user is not default.
In the scheme, a risk evaluation range is determined through comprehensive risk scores, if the risk is high, specific risk scores are analyzed, problematic assessment factors and risk scores thereof which are larger than or equal to a risk threshold are determined, if the number of problematic assessment factors is larger than the specified number, the user is judged to be an illegal user, and meanwhile problematic assessment factors and risk scores thereof are recorded for subsequent output display.
Preferably, S6 comprises the steps of:
s61: collecting an abnormal load diagram and extracting abnormal characteristic points;
s62: training an abnormal load graph, outputting the total number of abnormal loads, and marking the illegal electricity utilization behavior based on the total number of abnormal loads;
s63: and identifying the electricity utilization curve graph by adopting a target detection algorithm, and outputting the illegal electricity utilization behavior.
In the scheme, the abnormal load diagram is collected and trained to complete training of the data set, the relation between the total abnormal load and the illegal electricity consumption behavior is marked, the corresponding total abnormal load is obtained by taking the electricity consumption diagram as input, the illegal electricity consumption behavior is further obtained, and the total abnormal load and the illegal electricity consumption behavior can be automatically identified and matched according to the electricity consumption diagram.
A system for identifying a low voltage charging pile's default electrical behavior, comprising: the data acquisition module is connected with the data cleaning module, the data cleaning module is connected with the feature extraction module, the feature extraction module is connected with the model construction module, the model construction module is connected with the result judgment module, and the result judgment module is connected with the result display module.
In the scheme, the data acquisition module acquires historical load data of the charging pile through a Selenium technology and outputs original data. The data cleaning module performs cleaning processing such as denoising and abnormal value deleting on the original data and outputs cleaned data. The feature extraction module extracts a plurality of feature parameters from the cleaned data and outputs feature parameter data. The model construction module sets a risk score interval of each characteristic parameter according to the data of a plurality of characteristic parameters, obtains a comprehensive user power consumption risk score through multiplication of risk scores of the characteristic parameters, and outputs scoring data. The result judgment module outputs the user number of the offending user based on the comprehensive risk score and the risk score. The result display module reads and draws an original load data curve of the offending user within one week based on the user number of the offending user, outputs a power utilization curve graph, and judges the offending power utilization behavior based on the power utilization curve graph. The recognition accuracy of the illegal electricity utilization behavior of the low-voltage charging pile can be improved.
Preferably, the data acquisition module acquires load data of the charging pile through a browser automatic test framework technology, and acquires user data of the third party system through an interface.
Preferably, the method further comprises: the model optimization module is connected with the model construction module.
In the scheme, the model optimization module is used for inputting the on-site check feedback result, correcting the risk scoring interval and the comprehensive scoring threshold value of each characteristic parameter and realizing the parameter optimization of the model. And continuous optimization of the model is supported, historical data is integrated through a tool, so that the optimized data is more abundant, and continuous upgrading of the model is realized.
The beneficial effects of the invention are as follows: the recognition accuracy of the illegal electricity utilization behavior of the low-voltage charging pile can be improved.
Drawings
Fig. 1 is a schematic diagram of a system for identifying the default electricity utilization behavior of the low-voltage charging pile.
Fig. 2 is a flow chart of a method for identifying the default electricity utilization behavior of the low-voltage charging pile.
In the figure 1, a data acquisition module; 2. a data cleaning module; 3. a feature extraction module; 4. a model building module; 5. a result judging module; 6. a result display module; 7. and a model optimization module.
Detailed Description
The technical scheme of the invention is further specifically described below through examples and with reference to the accompanying drawings.
Examples: as shown in fig. 1, a system for identifying the default electricity utilization behavior of a low-voltage charging pile comprises:
and the data acquisition module 1 acquires historical load data of the charging pile through a Selenium technology and outputs original data.
The data cleaning module 2 performs cleaning processing such as denoising and abnormal value deleting on the original data, outputs cleaned data and is connected with the data acquisition module 1.
The feature extraction module 3 extracts a plurality of feature parameters from the cleaned data, outputs feature parameter data and is connected with the data cleaning module 2.
The model construction module 4 sets a risk score interval of each characteristic parameter according to the data of a plurality of characteristic parameters, obtains a comprehensive user power consumption risk score through multiplication of the risk scores of the characteristic parameters, outputs scoring data and is connected with the characteristic extraction module 3.
The result judging module 5 outputs the user number of the offending user based on the comprehensive risk score and the risk score, and connects with the model constructing module 4.
The result display module 6 reads and draws an original load data curve of the offending user within one week based on the user number of the offending user, outputs a power utilization curve graph, judges the offending power utilization behavior based on the power utilization curve graph, and is connected with the result judgment module 5.
And the data acquisition module 1 acquires historical load data of the charging pile through a Selenium (browser automatic test framework) technology and outputs original data. And operating a browser driver by using a Selenium technology to finish the derivation of the load data of the acquisition system. After the load data is obtained, user data of the third party system is obtained through an interface, the load data and the user data are spliced into original data, and then data linkage analysis processing is carried out, wherein the user data comprise user numbers, user names, user addresses, meter reading numbers, power supply units and the like.
The data cleaning module 2 performs cleaning processing such as denoising and abnormal value deletion on the original data, and outputs cleaned data. The program based on file traversal runs the core logic, an operation algorithm for directly calling the EXCEL bottom logic is applied, data processing is carried out on users with load data, the original data with the reject power smaller than the reject power threshold is rejected, in the embodiment, the reject power threshold is 20W, and missing data in a short time is filled with the latest non-zero sampling point data. In general, the common standby power of the low-voltage charging pile is smaller than 0.01kw, and the default power of the data preprocessing algorithm is set to be in a charging state when the default power is larger than the charging power threshold, in this embodiment, the charging power threshold is 0.02kw.
And the feature extraction module 3 extracts a plurality of feature parameters from the cleaned data and outputs feature parameter data. The characteristic parameters comprise average power, variance, peak-to-valley ratio, integral electric quantity, maximum load, minimum load, load mutation times, charging pile utilization rate and the like.
And the model construction module 4 obtains the risk scores of the assessment factors according to the data of the characteristic parameters, obtains the comprehensive risk score through multiplication of the risk scores of the assessment factors, and outputs the comprehensive risk score. And (3) establishing a charging pile characteristic parameter analysis model, and utilizing various characteristic parameters of the charging pile, selecting a proper mathematical model through classical algorithms such as decision trees, random forests, neural networks, convolutional neural networks and the like, and constructing the charging pile characteristic parameter analysis model. Constructing a risk assessment model, wherein assessment factors of the risk assessment model comprise a variance factor, a factor with more than 20w in 0-1KW, a factor with load segmentation number of '1 KW', an average power factor during charging, a charging pile utilization factor, an integral back calculation electric quantity factor and a capacity utilization factor, wherein the variance factor reflects the overall fluctuation condition during charging, and the larger the variance, the larger the fluctuation and the higher the risk; the higher the proportion of the above 20w duty factor in the 0-1KW is, the higher the risk is; the more the number of load segments is by a factor of ">1kw", the higher the risk is, in this example, 0.5 kw; the average power factor during charging is generally lower in risk of 6.5-7.5 kw, and has a unique distribution function; the higher the utilization factor of the charging pile is, the higher the risk is; integrating the back calculation electric quantity factor, wherein the higher the electric quantity is, the higher the risk is; capacity usage factor, has its unique distribution function. The product of the risk scores of all the assessment factors is the comprehensive risk score, and the larger the value is, the higher the risk is.
The result judging module 5 sets a risk evaluation range, wherein the risk evaluation range comprises no risk, low risk, medium and high risk, the comprehensive risk scores are ordered based on the risk evaluation range, for the high risk users, if the risk scores of a plurality of and more than one assessment factors are larger than or equal to a risk threshold value, the users are judged to be illegal users, corresponding user numbers are output, and otherwise, the users are judged to be no illegal users.
And the result display module 6 reads and draws an original load data curve of the offending user within one week based on the user number of the offending user, outputs a power utilization curve graph, judges the offending power utilization behavior based on the power utilization curve graph, and further carries out relevant processing.
Furthermore, it may further include: the model optimizing module 7, the model optimizing module 7 is connected with the model constructing module 4. The model optimization module 7 is used for inputting the on-site verification feedback result, correcting the risk score interval and the comprehensive score threshold value of each characteristic parameter, and realizing the parameter optimization of the model. And continuous optimization of the model is supported, historical data is integrated through a tool, so that the optimized data is more abundant, and continuous upgrading of the model is realized.
Collecting load data and user data, splicing the load data and the user data into original data, carrying out cleaning treatment such as denoising and abnormal value deleting on the original data, outputting the cleaned data, extracting a plurality of characteristic parameters based on the cleaned data, obtaining risk scores of all assessment factors based on the data of the plurality of characteristic parameters, obtaining comprehensive risk scores through the risk scores, judging the offending user based on the comprehensive risk scores and the risk scores, outputting the user number of the offending user, reading and drawing an original load data curve within one week of the offending user based on the user number of the offending user, outputting an electricity consumption curve, judging the offending electricity consumption behavior based on the electricity consumption curve, and further carrying out relevant treatment.
And the full flow from data acquisition to model optimization realizes the automatic analysis of the low-voltage charging pile default electricity utilization identification, and visual electricity utilization curve results are output. The logic carding and control of the complex technical process are realized, the modules are independent of each other but are mutually dependent, the transmission of data and information is completed in an input and output mode, and the systematicness and the integrity of the system are realized.
Embodiment two:
as shown in fig. 2, the method for identifying the default electricity consumption behavior of the low-voltage charging pile comprises the following steps:
s1: collecting original data;
s2: cleaning the original data to obtain cleaned data;
s3: extracting characteristic parameters based on the cleaned data, and setting assessment factors based on the characteristic parameters;
s4: obtaining a risk score of the assessment factor based on the characteristic parameters, and obtaining a comprehensive risk score based on the risk score;
s5: determining offending user information based on the risk score and the comprehensive risk score, and outputting a power utilization curve graph based on the offending user information;
s6: the default electrical behavior is identified based on the electrical usage graph.
In step S1, the browser driver is operated by using the Selenium technology to complete the derivation of the load data of the acquisition system. After the load data is obtained, the load data and the user data of the third party system are spliced into original data, and then data linkage analysis processing is carried out.
In step S2, the original data is subjected to cleaning processing such as denoising, filling of a null value, deletion of an abnormal value, and the like, and cleaned data is output. And carrying out data denoising processing on the user with the load data, wherein the rejection power is smaller than the primary data of the rejection power threshold, and in the embodiment, the rejection power threshold is 20W. The null value is filled, and missing data in a short time is filled with the latest non-zero sampling point data. In general, the common standby power of the low-voltage charging pile is smaller than 0.01kw, and the default power of the data preprocessing algorithm is set to be in a charging state when the default power is larger than the charging power threshold, in this embodiment, the charging power threshold is 0.02kw.
In step S3, the feature extraction module 3 extracts a plurality of feature parameters from the cleaned data, and outputs feature parameter data. The characteristic parameters comprise average power, variance, peak-to-valley ratio, integral electric quantity, maximum load, minimum load, charging pile utilization rate, load mutation times and the like. And (3) establishing a charging pile characteristic parameter analysis model, and utilizing various characteristic parameters of the charging pile, selecting a proper mathematical model through classical algorithms such as decision trees, random forests, neural networks, convolutional neural networks and the like, and constructing the charging pile characteristic parameter analysis model. The risk assessment model is constructed, and assessment factors of the risk assessment model comprise a variance factor, a duty factor of more than 20w in 0-1KW, a load segmentation number factor of '1 KW', an average power factor during charging, a charging pile utilization factor, an integral back calculation electric quantity factor and a capacity utilization factor.
In step S4, the characteristic parameters are input into the corresponding models, the risk scores of the assessment factors are output through the models, each assessment factor corresponds to one risk score, the comprehensive risk score of all the assessment factors is calculated, the product of the risk scores of all the assessment factors is the comprehensive risk score, and the larger the numerical value is, the higher the risk is.
In step S5, a risk assessment range is set, the risk assessment range includes no risk, low risk, medium and high risk, comprehensive risk scores are ordered based on the risk assessment range, for users with high risk, if there are several or more risk scores of the assessment factors greater than or equal to the risk threshold, it is determined that the users are default users, corresponding user numbers are output, otherwise, it is determined that no default exists. And reading and drawing an original load data curve of the offending user within one week based on the user number of the offending user, and outputting an electricity utilization curve.
In step S6, an abnormal load diagram is collected, and abnormal characteristic points are extracted; training an abnormal load graph, outputting the total number of abnormal loads, and marking the illegal electricity utilization behavior based on the total number of abnormal loads; and identifying the electricity utilization curve graph by adopting a target detection algorithm, and outputting the illegal electricity utilization behavior. And carrying out yolov7 identification on the load characteristic curve, carrying out AI automatic abnormality judgment, outputting an abnormality judgment result, namely the illegal electricity consumption behavior, and carrying out iterative sequencing on risk grades.
Embodiment III:
the embodiment provides a specific implementation scheme, which is based on the technology of acquiring and automatically cleaning the full load data of the low-voltage resident charging pile, and realizes the automatic derivation and cleaning of the power load data of the resident charging pile of 1.5 thousands of households. The measured 42.9G load data was derived at a rate of 462kb/s for 27 hours.
And carrying out data processing on the users with the load data, and eliminating the original data with the power less than 20W. The common standby power of the low-voltage charging pile is generally smaller than 0.01kw, and the charging state is set when the default power of the data preprocessing algorithm is larger than 0.02kw.
Setting assessment factors, wherein the assessment factors comprise more than 20w of duty factors in 0-1KW, load segmentation number factors of '1 KW', average power factor during charging, charging pile use rate factors, integral back calculation electric quantity factors and capacity use rate factors. The risk scores of the seven assessment factors range from 1 to 5 to evaluate the user's risk of electricity consumption against the contract. And obtaining the risk scores of the corresponding assessment factors through the model based on the characteristic parameters, and calculating the scoring comprehensive risk score of each user based on the risk scores of the assessment factors. Then, all users are divided into four categories, respectively: the comprehensive risk score of the risk is 1, the comprehensive risk score of the low risk is less than 10, the comprehensive risk score of the medium risk is more than or equal to 10 and less than 20, and the comprehensive risk score of the high risk is more than or equal to 20. For high risk users, if the risk scores of three or more factors are greater than or equal to 4, marking the users corresponding to the high risk comprehensive risk scores as default users, otherwise, judging that no default exists.
In order to accurately describe the load characteristic waveform of the charging pile, embody the electrical characteristics of the charging pile, facilitate identification and judgment, obtain characteristic parameters such as load size, load fluctuation regularity, load change trend and the like through data acquisition and processing, further analyze and mine load data by utilizing technologies such as data mining, yolov5 and the like, extract more effective characteristic parameters, construct abnormal evaluation indexes of the charging pile, and provide effective parameters for forming the load characteristic waveform of the charging pile.
And (3) collecting abnormal load characteristics, namely collecting 74 pieces of collected abnormal load graphs, wherein 7-20 abnormal characteristic points are collected for each piece, and 1082 abnormal characteristic points are collected in total. Training the abnormal load characteristics, namely training the abnormal load characteristics for 6 hours, wherein the iteration times are 100 times. After hundreds of iterations, the convergence and loss reach the target values. The abnormal load feature detection is carried out by adopting a model which is trained for 100 times in an iterative way, the confidence coefficient is set to be 0.4, the average detection time of a single picture is about 0.14 seconds, and the total number of detected abnormal loads is output and marked.
And according to the risk ranking of the charging pile default identification, deriving 7-day load data of the top-ranked user, and drawing a load characteristic curve (electricity utilization curve chart) of the top-ranked user. And (3) carrying out yolov7 (target detection algorithm) identification on the load characteristic curve, carrying out AI automatic abnormality judgment, outputting an abnormality judgment result, namely, the illegal electricity consumption behavior, and carrying out iterative sequencing on the risk level.
To verify the results, telephone confirmation and field inspection are performed on the offending user. The results show that there is actually a user percentage 82.16% of the offending electricity usage. Compared with the accuracy of 45% adopting a simple electric quantity threshold judgment mode, the method has the advantages that the recognition efficiency and accuracy of the default users are remarkably improved.
The accuracy of identifying the low-voltage charging pile against the default electricity utilization is improved, and more comprehensive and accurate judgment factors are considered by constructing a comprehensive evaluation model. The automatic data analysis of the low-voltage charging pile is realized, a system integrating data acquisition and analysis is developed, and the working efficiency is improved. And continuous optimization of the model is supported, historical data is integrated through a tool, so that the optimized data is more abundant, and continuous upgrading of the model is realized.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such alterations, combinations, and substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (10)
1. The method for identifying the illegal electricity consumption behavior of the low-voltage charging pile is characterized by comprising the following steps of:
s1: collecting original data;
s2: cleaning the original data to obtain cleaned data;
s3: extracting characteristic parameters based on the cleaned data, and setting assessment factors based on the characteristic parameters;
s4: obtaining a risk score of the assessment factor based on the characteristic parameters, and obtaining a comprehensive risk score based on the risk score;
s5: determining offending user information based on the risk score and the comprehensive risk score, and outputting a power utilization curve graph based on the offending user information;
s6: the default electrical behavior is identified based on the electrical usage graph.
2. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1, wherein in S1, the raw data includes load data and user data.
3. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1, wherein in S2, the cleaning process includes a denoising process, a null value filling process and an abnormal value deleting process.
4. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1, wherein in S3, the characteristic parameters comprise average power, variance, peak-to-valley ratio, integrated electric quantity, maximum load, minimum load, charging pile usage rate and load mutation times, and the evaluation factors comprise variance factors, duty factor, load segmentation number factors, average power factor during charging, charging pile usage rate factors, integrated back-calculation electric quantity factors and capacity usage rate factors.
5. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1, wherein in S4, the rating factors and the risk scores are in one-to-one correspondence, and the comprehensive risk score is equal to the product of the risk scores of part or all of the rating factors.
6. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1, wherein S5 comprises the following steps:
s51: setting a risk assessment range;
s52: judging whether the risk assessment range corresponding to the comprehensive risk score is high risk, if so, performing step S53; if not, judging that the user is not default;
s53: judging whether the risk scores of a plurality of and more than one assessment factors are larger than or equal to a risk threshold value; if yes, judging as a default user, outputting a user number of the default user, reading and drawing a load data curve within the appointed time of the default user based on the user number, and outputting an electricity utilization curve; if not, judging that the user is not default.
7. The method for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 1 or 6, wherein S6 comprises the following steps:
s61: collecting an abnormal load diagram and extracting abnormal characteristic points;
s62: training an abnormal load graph, outputting the total number of abnormal loads, and marking the illegal electricity utilization behavior based on the total number of abnormal loads;
s63: and identifying the electricity utilization curve graph by adopting a target detection algorithm, and outputting the illegal electricity utilization behavior.
8. The system for identifying the illegal electricity consumption behavior of the low-voltage charging pile is suitable for the method for identifying the illegal electricity consumption behavior of the low-voltage charging pile according to any one of claims 1 to 7, and is characterized by comprising the following steps: the data acquisition module is connected with the data cleaning module, the data cleaning module is connected with the feature extraction module, the feature extraction module is connected with the model construction module, the model construction module is connected with the result judgment module, and the result judgment module is connected with the result display module.
9. The system for identifying the default electricity consumption behavior of the low-voltage charging pile according to claim 8, wherein the data acquisition module acquires the load data of the charging pile through a browser automatic test framework technology and acquires the user data of the third party system through an interface.
10. The system for identifying the default electricity utilization behavior of the low-voltage charging pile according to claim 8 or 9, further comprising: the model optimization module is connected with the model construction module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310817174.6A CN116933201A (en) | 2023-07-05 | 2023-07-05 | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310817174.6A CN116933201A (en) | 2023-07-05 | 2023-07-05 | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116933201A true CN116933201A (en) | 2023-10-24 |
Family
ID=88378272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310817174.6A Pending CN116933201A (en) | 2023-07-05 | 2023-07-05 | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116933201A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118194055A (en) * | 2024-05-14 | 2024-06-14 | 国网江西省电力有限公司信息通信分公司 | Charging pile power curve matching method |
-
2023
- 2023-07-05 CN CN202310817174.6A patent/CN116933201A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118194055A (en) * | 2024-05-14 | 2024-06-14 | 国网江西省电力有限公司信息通信分公司 | Charging pile power curve matching method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106780121B (en) | Power consumption abnormity identification method based on power consumption load mode analysis | |
| CN112766550B (en) | Random forest-based power failure sensitive user prediction method, system, storage medium and computer equipment | |
| CN111738462B (en) | Fault first-aid repair active service early warning method for electric power metering device | |
| CN102521080B (en) | Computer data recovery method for electricity-consumption information collecting system for power consumers | |
| CN102692615B (en) | System capable of automatically acquiring electric quantity data | |
| CN109687447B (en) | Electric power energy consumption prediction method and device | |
| CN116933201A (en) | Method and system for identifying illegal electricity utilization behavior of low-voltage charging pile | |
| Xue et al. | Adaptive ultra-short-term wind power prediction based on risk assessment | |
| CN110968703B (en) | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm | |
| CN111179576B (en) | Power utilization information acquisition fault diagnosis method and system with inductive learning function | |
| CN111179108A (en) | Method and device for predicting power consumption | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN117485177A (en) | Intelligent energy-saving control system and method for charging pile | |
| CN115859099A (en) | Sample generation method and device, electronic equipment and storage medium | |
| CN115758151A (en) | Combined diagnosis model establishing method and photovoltaic module fault diagnosis method | |
| CN112327190B (en) | Method for identifying health state of energy storage battery | |
| CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
| CN116975389A (en) | Transformer substation fault analysis method and device | |
| CN116910655A (en) | Intelligent ammeter fault prediction method based on device measurement data | |
| CN116879672A (en) | Intelligent identification method for total fault of area based on time convolution network | |
| CN111614489A (en) | Transient recording fault data cleaning method and system based on sparse self-coding | |
| CN115528684A (en) | Ultra-short-term load prediction method and device and electronic equipment | |
| CN115511016A (en) | Incremental active learning-based electricity charge anomaly detection method and device | |
| CN114819248A (en) | Wind power generation capacity prediction method and device based on random forest | |
| CN113866562A (en) | Line loss identification method for power system distribution room |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |